PWM Inverter Drive explanation please

[Physicist3]

Hi, I understand that the PWM inverter drive system first uses a diode bridge rectifier in order to convert the AC incoming into DC before using an inverter to convert this DC into PWM for use on the induction motor. Could someone please explain in simple terms how the PWM inverter section works, (i.e. Pairing the wave with a triangle wave) etc and what the original sine wave and dc waves are used for?? Also, how does the mark-space ratio of the output signal change with a change in frequency?? How is the frequency changed?? Is it by changing the triangle wave frequency or what?? Thanks

 

[Windadct]

There are a few questions in your post that I'll try to straighten out..

I think you are mention "Original Sine" relative to the AC power TO the drive. The operation fof the VFD is independent of the sources AC.

The DC really should not be called a "wave" - ideally it is a perfectly flat DC source of power for the VFD Inverter. ( in the real world this is not perfectly flat - between 2-10% ripple - an undesirable result from both the Rectifier and the VFD inverter current)

OK - then the common way to generate the PWM Sine wave is a Triangle Wave operating at the Switching Frequency ( Fsw) and the width (modulation) of the pulse (output from the inverter) is determined by the the ratio of the triangle wave to the desired output (typically the Voltage). (That is not the best wording )

There are LOTS of very good references on this : http://www.ab.com/support/abdrives/documentation/techpapers/PWMDrives01.pdf

It looks like you are making a common mistake and tying to look at the whole system at once, the simpler answer is that the PWM wave is really using pulses to synthesize a wave- this can be done to make really any waveform from DC to just under 1/2 the FSW ( but then heavily distorted - high harmonics)

 

[Physicist3]

So am I right in thinking that the DC link output signal which goes to the inverter is 'chopped' using the IGBTs in order to reproduce an artificial sine wave? Is this then paired with the carrier triangle wave and then a PWM wave is created from the points of intersection between the two waves, with the PWM wave being 'on' in the area between two intersection points where the recreated sine wave passes over the triangle wave, and 'off' in the area where the wave passes under the point of one of the triangle points? How is the height of the PWM wave determined??

 

[Windadct]

Sorry for the delay - traveling for work. I (personally) really do not like referring to the DC as a signal, the word signal to me implies that it has some information or meaning, and DC in this case is JUST a power source. (in a DC analog feedback - this is a DC signal)

After that - yes, the Inverter chops in the manner you have described. The height of the PWM - is technically the full DC voltage, this is then filtered to develop an average of the ON (full V) and the Off (0 V)... so if the duty cycle of the output is 50%--- the average voltage is 1/2 of the DC link voltage.